Byhve VM crashing, how to troubleshoot?

[P-Chan]

So I was running 11.1-U5 (if memory serves), but was unable to get Ubuntu installed in Bhyve, I found others having issues with this online as well and switching to Nightlies worked for them. This isn't a production system, it's my personal system, so that was what I did. My server crashed partway through the upgrade to Nightlies, not entirely sure why, but I was left with a veritable mess that wasn't working. I ended up doing a fresh install of 11.1-U5, importing the encrypted disks (uploaded my GELI key), then restoring my configs. I then switched back to Nightlies - this time the upgrade seemed to work. With the latest Nightly build, I was able to get Ubuntu installed and running. I attached a ZVOL via iscsi and mounted an NFS filesystem and began transferring data between them (moving from a LUKS zvol to an rclone crypt volume). During the transfer, the VM crashed. FreeNAS was still running, but I couldn't start the VM anymore. I had to do a complete restart before it started. I tried a second time with similar failure. I decided to increase resources (from 1 vCPU and 2048 MB to 2 vCPUs and 4096 MB), but it still crashed. I had an SSH session open to the server when it crashed and it just loses connection. Where do I go to troubleshoot this? I've poked around several log files, but can't find anything amiss at first glance.

In the meantime, I will be updating nightly to see if my issue ever gets fixed, but if I can find some logs or something to help determine what the issue is, I would be happy to open a bug report. Thanks!

 

[P-Chan]

I found this in the logs:

Code:

[2018/06/10 19:46:14] (INFO) VMService.run():276 - ===> Error VM: Level-EE ID: 6 BHYVE_CODE: -10
[2018/06/10 19:46:14] (ERROR) VMService.running():398 - ===> VMM Level-EE is running without bhyve process.
[2018/06/10 19:46:14] (DEBUG) VMService.__teardown_guest_vmemory():298 - ===> Give back guest memory to ARC.: 4294967296
[2018/06/10 19:46:14] (WARNING) VMService.destroy_vm():281 - ===> Destroying VM: Level-EE ID: 6 BHYVE_CODE: -10
[2018/06/10 19:46:14] (DEBUG) VMService.kill_bhyve_web():365 - ==> Killing WEBVNC: 10622
[2018/06/10 20:14:18] (DEBUG) ServiceService._simplecmd():259 - Calling: restart(webshell)

Still not sure how this helps me, but it looks like the BHyve process ended and so the server decided to shutdown the VM.

 

[dlavigne]

Were you able to figure out the cause for this?